An Empirical Bayes Analysis of Photographic Traffic Enforcement Systems in Texas

In Texas, there were over 11,600 crashes in 2011 associated with Red-Light Related (RLR) violations. To improve intersection safety related to RLR violations, automated photographic traffic signal enforcement systems, also known as red-light running camera (RLC), have been installed at signalized intersections. In 2011, 50 communities reported operating RLC systems at 398 intersections.

The primary objective of this research was to evaluate the safety effectiveness of RLC systems using the before-after study with the Empirical Bayesian (EB) method in Texas. A before-after study using a naïve, comparison group (CG), or EB method can be used to evaluate the safety effectiveness of RLC systems. However, naïve and CG methods suffer from the important limitation of regression-to-mean (RTM) bias. The EB method was used in this study to address the RTM bias during the evaluation of treatments. Using the EB method allowed TTI to estimate safety benefits at treated sites based upon reference sites with similar traits and without RLC treatments. Using the EB method, this research results indicated significant decrease in all type and right-angle RLR crashes by 20% and 24%, respectively, while a significant increase in rear-end RLR crashes by 37%. This result is consistent with the findings from the naïve and CG methods. Thus, irrespective of the method used, the RLC systems have shown to reduce all type and RA crashes related to RLR violations, while an increase in RE crashes.

The secondary objective of this research was to analyze the criteria used for selecting the intersections for RLC treatment placement. In terms of site selection criteria, results suggested that if intersections experienced less than two RLR crashes per year or one RLR crash per 10,000 vehicles per year are treated then one can expect an increase in all type and RA RLR crashes. If the intersections have four or more RLR crashes per year or two RLR crashes per 10,000 vehicles per year, then the treatment will significantly decreases all type and RA crashes.

Get the complete facts about the Empirical Bayes Analysis of Photographic Traffic Enforcement Systems in Texas by downloading the PDF below.

 

 

December 26th, 2012|